| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ORIGINAL ARTICLES |
From the Department of Psychiatry, Stanford University and VA Palo Alto Health Science Center, Palo Alto, CA (T.R.) and Department of Psychology, St. George’s Hospital Medical School, University of London, London, UK (A.S.).
Address reprint requests to: Thomas Ritz, PhD, VA Palo Alto HSC (116F PAD), Dept. of Psychiatry, 3801 Miranda Avenue, Palo Alto, CA 94304. Email: tritz{at}stanford.edu
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESACKNOWLEDGMENTSREFERENCES |
|---|
METHODS: Twenty asthmatic patients and 20 nonasthmatic control subjects participated in a laboratory session in which various emotions (ie, anxiety, anger, depression, happiness, elation, contentment, and neutrality) were induced by films. Respiratory resistance (Ros) was measured by forced oscillation. After this session, participants kept mood diaries, including regular spirometric self-assessments, for at least 3 weeks. Episodes of strong negative or positive mood were selected from these diaries and compared with conditions of relative affective neutrality.
RESULTS: In asthmatic patients, negative mood states, and to a lesser degree positive mood states, were associated with a reduction in forced expiratory volume in the first second (FEV1) compared with neutral states. These effects were not observed in nonasthmatic control subjects. Self-reports of arousal varied in a reciprocal manner with FEV1, whereas physical activity did not vary systematically between mood episodes. A moderate negative relationship between changes in FEV1 during negative mood episodes and changes in Ros during viewing of the depressing film was also observed in asthmatic patients.
CONCLUSION: Pulmonary function of asthmatic patients is negatively affected by strong mood states in daily life. Airway effects of negative emotion induction, particularly depression, can predict changes in pulmonary function in response to negative mood in the field.
Key Words: emotion, • asthma • respiratory resistance, • lung function diaries • laboratory-field comparison.
Abbreviations: ANOVA = analysis of variance; FEV1 = forced expiratory volume in the first second; PEF = peak expiratory flow; Ros = oscillatory resistance.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESACKNOWLEDGMENTSREFERENCES |
|---|
To examine the predictive validity of laboratory measurements for real-life performance, a combination of laboratory and field assessments is usually used (13). Comparable research has not yet been undertaken to determine the predictive validity of airway responses to behavioral tasks. In addition, only a few field studies have investigated the relationship between affective states and airway obstruction under more naturalistic conditions. Typically, paper-and-pencil diaries of daily mood states have been correlated with self-assessments of lung function by miniature peak-flow meters. Evidence from this type of study remains inconclusive because idiosyncratic associations of mood and lung function changes were usually observed (14–17; however, see Ref. 18).
We recently studied asthmatic patients and nonasthmatic control subjects in a laboratory investigation in which various emotions were experimentally induced by showing affective film clips (19). After the laboratory assessments, the majority of participants completed a 3-week diary of their mood, activity, and lung function. This field-assessment phase was designed to study the relationship between mood and lung function using a combined electronic diary and spirometry method, which guaranteed higher validity than the common paper-and-pencil method and included FEV1, a more sensitive parameter of lung function. In this report, we examine the relationship between airway responses during emotion induction in the laboratory and during particularly strong mood states in the everyday life of participants. For this purpose, we selected lung function measurements from the diary-assessment phase at times of extreme positive or negative mood states and for states of relative affective neutrality. We were interested in whether these measurements would replicate the pattern of changes in airway resistance observed in the laboratory, which showed increases in resistance during viewing of positive and negative emotional films compared with a neutral film clip. We expected this pattern to be matched by arousal ratings in the diary, which would suggest that valence-nonspecific emotional arousal was associated with a reduction in lung function. In addition, we sought to determine whether airway reactivity during viewing of affective films was predictive of changes in lung function during strong emotional states in everyday life.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESACKNOWLEDGMENTSREFERENCES |
|---|
All participants were nonsmokers and were free of psychiatric illnesses, cardiovascular diseases, family history of cardiovascular diseases, and respiratory diseases other than asthma. Nonasthmatic subjects were chosen at random from the general practice database and were matched on the basis of age and sex. Asthmatic patients had a mild or moderate degree of illness severity (20). They continued taking their prescribed medication (mainly ß-adrenergic bronchodilators and inhaled corticosteroids). Patients using oral corticosteroids or long-acting bronchodilators were excluded. Nine asthmatic patients used their ß-bronchodilators regularly throughout the diary period. For the laboratory appointment, patients were asked to take their last dose of bronchodilator (if necessary) at least 8 hours before arriving. If a patient had difficulty tolerating withdrawal from medication or if symptoms became aggravated, the appointment was rescheduled. In a background questionnaire, 11 patients stated that "psychological factors" (eg, worry, stress, or emotion) could cause their asthma symptoms, 6 stated that this was only true in combination with other trigger factors, and 3 ruled out a role of psychological factors.
Laboratory Testing: Emotional Film Sequences
Participants viewed film sequences that had been preevaluated for eliciting different emotional states; most sequences were taken from commercial movies (21). Seven sequences were selected to induce anxiety (226 seconds), anger (245 seconds), depression (258 seconds), elation (276 seconds), happiness (290 seconds), contentment (90 seconds), or a neutral state (180 seconds). After pilot testing, we chose two sketches of a British comedian to induce happiness and elation to match cultural preferences. The film clips were shown in random order.
Equipment and Measures
Total respiratory resistance was measured by using the forced oscillation technique (22). A Siemens Siregnost FD 5 with a fixed oscillation frequency of 10 Hz was used (23). Participants breathed ambient air through a mouthpiece and tube (resistance = 0.015 kPa/liter per s) with the nose occluded. An elastic band supported the cheeks and base of the mouth to reduce shunt characteristics of the tissue. The obtained resistance measure (Ros) is expressed in kilopascals per liter per second. Good correspondence with the most sensitive measurement technique of airway resistance, body plethysmography, has been reported (24, 25). The signal was recorded in a moving average of 7 seconds using a Grass 7D polygraph (Stag Instruments, Chalgrove, UK). Protocols were manually scored offline after artifacts in Ros due to swallowing were eliminated. A rater blinded to the experimental conditions scored half of the protocols, and no differences in reactivity scores due to raters were found.
PEF and FEV1 were measured with an electronic pocket spirometer (Jaeger Toennies AM2 asthma monitor) set up to correspond with standards of the European Respiratory Society (26). The instrument incorporated diary facilities with a microcomputer display for questions preprogrammed by the researcher. Diary entries and measurement results were stored in the electronic memory of the device, which allows direct recording of patient compliance in contrast to the commonly used paper-and-pencil diaries (27, 28). A list of 12 questions about symptom experience, mood, and activities preceded measurements of lung function at specified times in the morning, afternoon, and evening. From this list, the following questions on current mood are relevant for this report: 1) "Do you feel very aroused, aroused, calm, or very calm?" 2) "Do you feel content?" 3) "...anxious?" 4) "...depressed?" 5) "...elated?" 6) "...angry?" 7) "...happy?" Responses were "not at all," "slightly," "moderately," or "very" for each of the items 2 to 7. The answer categories for the arousal item were explained by the instructions for the Self-Assessment Manikin (29). In addition, participants were asked about their level of physical activity during the previous 30 minutes (none, mild, moderate, or heavy). The following examples were provided: none: lying down, sitting still, or standing still; mild: normal walking or light manual work at a desk (eg, writing or typing); moderate: fast walking, climbing a few flights of stairs, or lifting or carrying heavy items; or heavy: running (eg, for a bus or train), fast uphill walking, climbing several flights of stairs, lifting or carrying several heavy items, dancing, physical exercise, or sports. Another menu entry served as an event marker for indicating use of the bronchodilator. Asthmatic patients were asked to record each dose immediately after intake.
The instrument was preprogrammed to present the list of 12 questions on activation before any measurement could be taken. Once the questions had been answered, they were not presented until the next measurement period. The event marker for use of medication was available at all times. The highest measurement within the 10-minute period after the question series was saved in the electronic memory. Participants were instructed to take two measurements in a row or to continue with additional measurements if they felt that they had not invested the maximum possible effort into their expiratory maneuver.
Procedure
Participants arrived for the laboratory session between 3:30 and 7:00 PM. Reference to the effects of emotions on the airways was avoided in the background information and instructions. Full information on the aims of the study was given only at the end of the diary period. Patients signed the consent form and were trained to use the forced oscillation device and electronic spirometer. Measurements for the laboratory protocol were made in a light- and temperature-controlled room. The data-recording and video devices were controlled by the experimenter from an adjacent room. Participants were seated in a reclining chair in the upright position and watched the film clips on a television screen (53 x 39 cm) placed approximately 1.5 m in front of them. The protocol started with a 5-minute baseline period, after which the seven film sequences were shown in random order. Measurements were interrupted after each film presentation, and participants were allowed to remove the mouthpiece and nose clip. Two additional laboratory tasks were administered before or after the seven film clips (not discussed in this report), with the order of films and tasks counterbalanced within both subject groups. At the end of the laboratory session, participants were invited to return for the diary phase of the study. During the second appointment, patients were given instructions for use of the diary functions of the AM2 and retrained to use the forced expiratory maneuver if necessary. All menu entries of the diary display were explained in detail. Written instructions were provided with the instrument. Participants were instructed to take measurements three times per day for 21 consecutive days at the following times: between 8:00 and 10:00 AM (morning measurement), between 2:00 and 4:00 PM (afternoon), and between 8:00 and 10:00 PM (evening). They were encouraged to take a few practice days and to call the experimenter if they encountered any problems with the device. After 1 week, the experimenter called the participants to find out whether any problems had occurred.
Data Reduction and Analysis
For airway reactivity during film viewing, difference scores were calculated by subtracting Ros during the neutral film from Ros during the respective emotional film. Because lung function measurements are usually dependent on the circadian rhythm (30, 31), we chose the afternoon and early evening measurements as the most suitable data set for the laboratory-field comparison. (The complete data set from the diary assessment is the subject of a separate report.) Because a number of measurements did not fall into the designated 2-hour afternoon measurement period between 2:00 and 4:00 PM, the present analysis was based on a wider afternoon/early evening category comprising measurements made between noon and 8:00 PM. On average, 21 measurements were available for this time of the day across participants, with a range of 9 to 39 measurements in individual participants. This wide range was due to some participants missing one or more measurements per day and others continuing the diary beyond the required period.
Scores of 0 ("not at all") to 3 ("very") were assigned to the four answer categories of diary mood questions. For ease of interpretation, the scoring direction for the arousal question was reversed, with 0 for "very calm" and 3 for "very aroused." Because variability of the diary questions for some mood states was low or even zero in some participants, composite scores of negative mood (sum of ratings for anxious, depressed, and angry) and positive mood (sum of ratings for happy, elated, and content) were calculated. Although differential airway responses could be expected theoretically for different negative emotional states (9), overwhelming clinical and experimental evidence indicates a rather uniform response pattern of increases in airway resistance in response to various types of emotional arousal (2–12, 20). Positive and negative scores were negatively correlated within participants (mean r = -0.55). To determine episodes of strong positive, strong negative, and neutral mood during the diary period, a bipolar index of mood balance was then calculated by subtracting the negative from the positive score. This score ranged from -9 to +9 for maximally negative and positive mood, respectively. In accord with a dimensional concept of affect (32), the zero point of the scale was assumed to be representative of a neutral affectivity. For each patient, all lung function measurements at the most extreme positive or negative category on the bipolar scale as well as at (or nearest to) the zero category (for neutral episodes) were selected and averaged across the diary period. Reactivity scores were then calculated by subtracting the mean of the neutral category from the mean of the most extreme positive and negative category. Bronchodilator use was coded in a dummy variable assigning values decreasing from 6 to 0 for the first hour to the seventh hour after use of the inhaler and zero to any other hour at a further distance in time. This is in line with the average duration of most of the commonly available short-acting bronchodilators (33). For patients who did not use a bronchodilator, the dummy variable was set to zero.
To confirm differences between the diary entries selected for strong positive, strong negative, and neutral mood episodes, the index of mood level as well as the positive and negative mood composite scores (averaged for the selected episodes) were submitted to separate two-way repeated-measures ANOVAs with group (asthmatic or nonasthmatic) as the between-individual variable and episode (negative, neutral, or positive) as the within-individual variable. Because the zero score on a bipolar mood balance scale could also represent situations of high emotional ambiguity (34), post hoc tests (Newman-Keuls, p < .01) were used to explore whether positive and negative mood composite scores were lower for the neutral than the emotional conditions. Similar ANOVAs were performed for self-reports of arousal and physical activity, and for FEV1 and PEF to study their variation as a function of mood episodes during the diary period. Significance levels corrected using the Greenhouse-Geisser method are reported where necessary. From the results of the laboratory part of the study and earlier research (8, 19), we expected decreases in PEF and FEV1 in association with increases in arousal during positive and negative mood relative to neutral mood. This was tested by an a priori quadratic contrast across negative, neutral, and positive mood episodes (35). To test the predictive validity of laboratory induction of emotion, we calculated rank correlations (Spearman’s 
) between reactivity scores for film viewing in the laboratory and reactivity scores for positive or negative episodes in the field. Significant negative correlations (p < .05, one-tailed) were expected between Ros and changes in FEV1 or PEF, indicating that patients with a stronger increase in Ros in the laboratory also showed a stronger decrease in lung function in the field. To control for the effects of medication in asthmatic patients, field reactivity scores were based on the residuals from the regressions of FEV1 on the medication dummy variable.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESACKNOWLEDGMENTSREFERENCES |
|---|
|
Selection of Mood Episodes for the Field Study
Means of mood balance and composite scores of positive and negative mood based on the individual extremes of positive, negative, and neutral are shown in Table 2. The main effects for mood balance and positive and negative mood were highly significant (F(2,76) = 227.30, 260.28, and 63.78, respectively; p values < .001). Negative mood states did not reach the same extreme level as did positive states. Although neutral mood was characterized as a mixture of mild negative and positive mood, post hoc tests showed that the scores were significantly lower than positive mood during the positive episodes and negative mood during the negative episodes. Thus, the neutral midpoint of the mood balance scale did not reflect a high level of emotional ambiguity. Finally, mood episodes did not differ in terms of self-reported physical activity (F 1).
|
15% during negative episodes and 4 had such decreases during positive episodes, whereas only 1 control subject met this criterion during negative episodes. Self-reports of emotional triggers were not related to changes in FEV1 (Kruskal-Wallis test). No significant effects were found with PEF.
|
Laboratory-Field Correlation of Reactivity Scores
Because of missing effects of mood on PEF in the field, the laboratory-field correlations were calculated only for FEV1. Table 3 shows correlations between increases in Ros during viewing of the different emotional films and changes in FEV1 during episodes of strong positive or negative mood. Coefficients were low in general and insignificant for nonasthmatic subjects. For asthmatic patients, deterioration of lung function during negative mood in the field was correlated significantly with increases in Ros in the laboratory during the depressing film and the average for all negative films (r(20) = -0.55 and -0.38, respectively; p values = .006 and .050, respectively).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESACKNOWLEDGMENTSREFERENCES |
|---|
Mean changes in lung function were reflected in a similar pattern of self-reported emotional arousal for both groups of participants. Although the agreement was not perfect (due to a lack of significance or trend toward significance for the Group-by-Episode interaction effect for arousal), it converges with earlier laboratory findings, which showed comparable patterns for self-reported emotional arousal and airway resistance (8, 19). Greater physical activity before strong emotional episodes had not led to this specific response pattern. Heavier physical activity is known to elicit airway obstruction in asthma, typically after cessation of the activity (38). Although we did not measure physical activity directly, self-reported data can reflect actual activity quite well in ambulatory monitoring studies (39).
In earlier studies on daily mood and PEF (14, 17), highly idiosyncratic and situation-dependent associations between mood ratings and lung function were observed, suggesting zero relationships in average within-participants correlations. In contrast, the present analysis of extreme mood states converged on a more distinctive pattern of changes in lung function. A certain degree of nonlinearity might be inherent in the mood-spirometry relationship, whereby only extreme mood states are informative from a psychological point of view. This could be due to a low sensitivity of the measurement technique: Only larger lung function changes in extreme mood states might be captured reliably, whereas a majority of measurements during mild mood states would be governed by other factors or just contribute to error variance. Depending on the eventfulness of the patients’ life during the diary period, extreme mood states might be represented to a different degree in the individual diaries, thereby creating an impression of idiosyncrasy. A lower sensitivity of spirometry to mild changes in airway resistance has been observed in comparisons with body plethysmography (40) or forced oscillation (41).
A factor likely to contribute to this low sensitivity is the maximal expiratory maneuver, which requires inspiration to total lung capacity. Deep inspiratory maneuvers relax the smooth muscles of the airways (42), and pharmacologically induced increases in airway tone can be reduced by deep inspiration (36). On average, this might have led to a more conservative assessment of mood-related lung function changes: The mean percentage FEV1 difference between negative and neutral episodes was well below the criterion of 
FEV1 15%, which is commonly taken as the lowest threshold for clinical significance (1). Five asthmatic patients (25% of the asthmatic group) responded above this threshold during negative episodes, and only two patients reached a more stringent criterion of FEV1 20%.
Restriction of our sample to patients with mild or moderate asthma could also have contributed to smaller mood-related airway responses. On the other hand, it could be speculated whether our sample actually overestimated the mood-lung function relationship: More than half of the asthmatic patients reported a clear role for psychological triggers of their asthma. However, this is not an unusual percentage, because reports of psychological factors as the dominant triggers of asthma range from 4 to 70% (11). A number of methodological issues, such as differences in instruments and assessment settings, can account for this variation. The importance of these reports should therefore not be overestimated. Clearly, a psychometrically sound approach to the assessment of perceived asthma triggers is needed.
The correlations between laboratory and field reactivity were not spectacular. Only changes in lung function during the depressing film showed a significant relationship with negative mood episodes in asthmatic patients. The lack of a relationship between positive mood episodes and any of the positive films is not surprising given the comparably small changes in lung function. The outcome is still acceptable considering the number of factors that potentially acted against the demonstration of a laboratory-field relationship in this study. First, two different measurement techniques of pulmonary function were employed with different sensitivity for mild changes in airway tone. Unfortunately, more sensitive and direct techniques of lung function measurement are not feasible for use in the field. Including spirometry in the laboratory protocol with repeated measurements in close succession was contraindicated because irritation of the airways by the forced expiratory maneuver is known to produce bronchoconstriction in some asthmatic patients (43). Second, scheduling problems due to a restricted number of electronic spirometers led to a time gap between laboratory and field assessments of several weeks on average. Long-term stability of Ros difference scores has been shown to be quite low even for basic laboratory tasks (44). Given these obstacles, and the fact that laboratory-field associations of cardiovascular reactivity are also only modest on average (45), we interpret the present finding as modest evidence of the predictive validity of laboratory emotion induction for mood-related decline in lung function in the everyday life of asthmatic patients.
The laboratory-field association shown specifically for the depression film is consistent with a considerable amount of literature that suggests a specific role for sadness, depression, or behavioral states of withdrawal in asthma. The theme of earlier psychodynamic theorists’ strongly debated interpretation of the asthmatic response as suppressed crying (46; for review, see Ref. 11) was echoed in scenes of our depression film depicting a boy crying about the death of his father. Knapp et al. (47–49) identified fear of loss and depression as central psychological characteristics of asthma and interpreted the asthmatic attack as a physiological manifestation of a massive behavioral inhibition. A potential role of depression-like states in childhood asthma has been highlighted more recently by Miller et al. (50–52). A common characteristic of these emotional states is thought to be a strong parasympathetic discharge. Because the airways are primarily constricted by vagal excitation (53), the confrontation with depressing material can be expected to be a particularly powerful stimulus. Even if the etiological model underlying some of the older psychosomatic approaches can be rejected (11, 54), a particular susceptibility of sensitized or hyperresponsive airways to this type of stimulus would be anticipated. Depression in general also seemed to have been of particular importance in our asthmatic group. In an initial analysis of the diaries across the entire day we found a higher level of depressed mood in asthmatic patients than in control subjects (55). However, an analysis of reactivity scores for depression rather than general negative mood in the laboratory-field comparison was not feasible because of the lack of variability of afternoon depression ratings in a number of patients.
In conclusion, episodes of strong negative mood are associated with reduced lung function in the everyday life of asthmatic patients. These changes rarely reach common criteria of clinical significance when measured by spirometry. However, decreases in lung function during negative mood episodes are related to similar airway responses during negative emotion induction in the laboratory, specifically during induction of depressed affect by film viewing. It remains to be established whether more sensitive measurement techniques yield, and whether patients with more severe asthma have, stronger airway responses during everyday life.
|
ACKNOWLEDGMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESACKNOWLEDGMENTSREFERENCES |
|---|
|
NOTES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESACKNOWLEDGMENTSREFERENCES |
|---|
Received for publication February 7, 2000.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESACKNOWLEDGMENTSREFERENCES |
|---|
This article has been cited by other articles:
 |
|
 |
|
  A. Kullowatz, D. Rosenfield, B. Dahme, H. Magnussen, F. Kanniess, and T. Ritz Stress Effects on Lung Function in Asthma are Mediated by Changes in Airway Inflammation Psychosom Med, May 1, 2008; 70(4): 468 - 475. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Ritz, A. Steptoe, C. Bobb, A. H. S. Harris, and M. Edwards The Asthma Trigger Inventory: Validation of a Questionnaire for Perceived Triggers of Asthma Psychosom Med, November 1, 2006; 68(6): 956 - 965. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P Lehrer Anger, stress, dysregulation produces wear and tear on the lung. Thorax, October 1, 2006; 61(10): 833 - 834. [Full Text] [PDF]
|
|
|
|
|
|
J. M. Feldman, M. I. Siddique, E. Morales, B. Kaminski, S.-E. Lu, and P. M. Lehrer Psychiatric Disorders and Asthma Outcomes Among High-Risk Inner-City Patients Psychosom Med, November 1, 2005; 67(6): 989 - 996. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. E. Meuret, F. H. Wilhelm, and W. T. Roth Respiratory Biofeedback-Assisted Therapy in Panic Disorder Behav Modif, September 1, 2001; 25(4): 584 - 605. [Abstract] [PDF]
|
|
|
|
|
|
T. Ritz Relaxation Therapy in Adult Asthma: Is There New Evidence for Its Effectiveness? Behav Modif, September 1, 2001; 25(4): 640 - 666. [Abstract] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
